Update a ton of audio code to use size_t more correctly and in general reduce

webrtc/modules/audio_processing/voice_detection_impl.cc

Index: webrtc/modules/audio_processing/voice_detection_impl.cc
diff --git a/webrtc/modules/audio_processing/voice_detection_impl.cc b/webrtc/modules/audio_processing/voice_detection_impl.cc
index 0883536d52d39151d3fbe0f6d9dd2fbeb43f3511..710df4233fa55bab5246feba1309519529743c14 100644
--- a/webrtc/modules/audio_processing/voice_detection_impl.cc
+++ b/webrtc/modules/audio_processing/voice_detection_impl.cc
@@ -140,8 +140,8 @@ int VoiceDetectionImpl::Initialize() {
   }
 
   using_external_vad_ = false;
-  frame_size_samples_ = frame_size_ms_ *
-      apm_->proc_split_sample_rate_hz() / 1000;
+  frame_size_samples_ = static_cast<size_t>(
+      frame_size_ms_ * apm_->proc_split_sample_rate_hz() / 1000);
   // TODO(ajm): intialize frame buffer here.
 
   return apm_->kNoError;